Good afternoon. We are here today with Professor Neil Ferguson, Director of J-IDEA, talking about the current coronavirus outbreak. Welcome. Thank you for taking your time to speak with us. Would you be able to give us an update on the current estimates that you're working on with your team. So the main things we're working on currently are trying to get a better handle on really how lethal this virus is, what threat it poses to the human population. It's a difficult thing to do in an epidemic because just simply looking at how many people died so far, versus how many cases there are, tells you very little. It takes a quite long time often for somebody from being diagnosed, as a case to actually know whether we know they're going to live or survive the infection. So we're trying to correct for those factors. In terms of looking at the overall scale of the epidemic, we're going to confirm our estimates. What we can say is surveillance ongoing in China is really probably picking up if you think of all infections out there being a pyramid, it picks up the top of the pyramid. The most severe infection, we think probably 10 percent or less of all infections in China are being detected at the current time. The next tier down is really what's being detected overseas. We think sensitivity is somewhat higher, but still we may be only detecting maybe a quarter of all infections at that level. Lots of people will be entering that the borders are porous. Countries won't be detecting every case coming in. Would you be able to say anything about forward predictions of where the current outbreak is going? Well, in terms of where we think the current outbreak has got to in terms of scale, we estimate that maybe up to 50,000 new infections a day occur in China. Just obviously much larger than the official case numbers. It's going up all the time. So the forward projections depend really on the effectiveness of control measures. We think the epidemic is roughly doubling in size about every five days at the current time. It's hard to evaluate how effective controls are, but there's limited evidence of it's slowing in China. Under that scenario, the epidemic would really follow a natural cause, probably peak in its epicenter, Wuhan in around a month's time, some uncertainty around that, and then maybe a month or two later in the whole of China. The rest of the world will see epidemics at various times off to that, depending on how well connected they are, have been in the past few weeks. How many travelers have come to them are infected from China itself. As to the overall effectiveness of control measures, it's hard to evaluate if there's a lot of community transmission going on. We think there probably is, will be very hard to control this epidemic in the same way as, for instance, we control the SARS epidemic some 15, 20 years ago. Would you be able to explain how the mathematical modeling that your team is doing can inform policy and government response? Yes, so statistical and mathematical analysis of epidemic data has quite a long history. It's really become quite established within the policy making framework used particularly in the UK, but I think further field, and that's because often what we have, it's very noisy and limited data in an outbreak. We want to be able to get the most information from that possible. So advanced, statistical modeling and then later mechanistic transmission modeling can be useful in doing that. So simple questions we tried to answer is, how transmissible is a new pathogen? How long has it been circulating? How much occasion cases are always underestimated in every epidemics, and surveillance is never 100 percent perfect? So how large is the real epidemic? Then particularly important to how people respond is how severe a threat is posed to human health. Would you be able to explain a little bit about the uncertainty ranges around the estimates that you report? So uncertainty is always high early in a epidemic, particularly of a disease we've never seen before. We don't know critical things like how long it takes from when somebody's infected to when they develop symptoms. Something called incubation period. We don't have transmissible diseases. So we try to estimate these things from quite limited data, and the limited data means that there's always uncertainty in the estimates, which we tried to characterize using state of the art statistical methods. So often we will quote something like, we think every case generates 2.6 over cases on average, something called the reproduction number. But there'll be a range around that, might vary from anywhere between 2.1 up to 3.3, or something that captures how much we know about that information, that particular parameter of interest. Then with that uncertainty range, how useful is that to inform policies and government? So generally, policymakers would like everything to be certain. So handling uncertainty is a challenge. I think the uncertainty range tells them what confidence we have in the estimate if we've calculated correctly. So if we, for instance, say, well, the case fatality ratio of proportion people who might die in an outbreak varies. Could be anywhere from one in 1,000 to one in 10. That's not terribly informative estimate. If we can say precisely the estimate is one percent case fatality, one percent of people will die, and it could be 0.5 percent, it could be two percent. That's a much more useful estimate. Do you have any suggestions as to what the most efficient interventions would be at the moment for either China or internationally? Unfortunately, for this virus, we have limited options for interventions. At the current time, we have no vaccine. Their vaccines are rapidly being developed and it would take months, if not years to develop. Some testing of anti-viral treatments, existing drugs been re-purposed to see if they work in infections in these cases. The other measures we have to draw upon in what are called public health intervention. So identifying as many cases as possible, as early as possible and their disease, and isolating them is how we managed to control, for instance, the SARS outbreak, genetically quite similar virus to this new virus. Whether those measures will be as effective in this case remains to be seen. This virus has a much wider range of severity of symptoms. Quite a lot of people being infected may just have quite mild respiratory disease, might not even seek health care for it. If those people are transmitting, then it will be very hard to stop transmission overall. So next few weeks will really tell us about the likelihood of control measures working. So in terms of interventions, Chinese had been throwing everything they can at this outbreak for the last two weeks. In terms of what might work, that critically depends on who is transmitting infection and whether the people who are responsible but most of them are transmission, are being identified and isolated on early enough. The Chinese are currently focusing on identifying severe cases and isolating those. It's unclear whether the mild cases left in the community, and we think that's the majority of cases are still sustaining the outbreak or not. We'll know more in the next couple of weeks. At the moment, we can't see much evidence of the epidemic slowing down. There have been few fatalities outside mainland China. Would you be able to explain why we see the difference with the fatalities that we see in China? So looking at the severity of an epidemic, working out how many people might eventually die from it is very challenging. There were delays, quite often long delays. When somebody is diagnosed with an infection or may be reported in the media, for instance in Thailand or Japan or Hong Kong. When we finally know what the outcome of that case is, people will persevere, cases will be in intensive care maybe for weeks. We think the delay from when somebody is developed symptoms to when they might die is as long as 20 days or longer, three weeks or so. So the fact we haven't seen many deaths in travelers, we've seen two so far, cases detected outside mainland China, is not terribly reassuring. We wouldn't have expected to see very many so far because of those delays in the system. What we're trying to do now is account for those delays and do an accurate estimate of how lethal we think this virus is. How worried are you about this outbreak? The outbreak is clearly of concern. It's probably the most major, we call it emerging infectious disease event the world has seen at least since the pandemic of influenza in 2009 and maybe since the SARS outbreak in 2003. So we don't fully know enough to be able to estimate the full scale of disease burden we call it, which will affect the human population, how many people will die? But it's something that certainly governments do to prepare for urgently in terms of ramping up health system preparedness. In particular, undertaking rapid research to try to develop treatments which can reduce potentially fatality in severe case the disease. How is the work that you do funded? So the funding for the work we do here comes from a number of sources. We are very fortunate to have generous institute funding from Community Jameel to fund our work. Not just on responding to outbreaks, but much broader work on how the world responds to other forms of health crises. Teach climate change and the like. We then receive a lot of research funding from individual research funders. So within J-IDEA, we have the MRC Center for Global Infectious Disease Analysis, which has generous funding from the UK Medical Research Council, receivables of funding from other funders as well. Thank you for taking the time to speak to us today. It's a pleasure.
